Every counter reports a number per millilitre and they do not mean the same thing. A brightfield instrument finds objects of about the right size and shape; a fluorescence instrument counts things that took up a dye, which is a different set; an impedance counter counts what displaced enough electrolyte. Put one suspension through three of them and the answers differ by more than anybody expects, and none of them is wrong. Choosing well is a matter of deciding which definition of a cell your experiment needs, and then holding one instrument and one protocol for the whole study.
Where a fluorescent cell counter earns the reagent
Dye based counting wins where the sample is not a clean suspension of one cell type. Debris, red cells, platelets and dead nuclei all confuse a brightfield algorithm and none of them takes up a live cell dye, so a primary preparation or a tissue digest counts far more reliably with a stain than without one. The cost is a reagent per sample and an incubation, which is why nobody uses it for routine passaging of a cell line and everybody should use it for a primary isolate.
What fluorescent cell counting adds over brightfield, and where fluorescence cell counting stops
It adds a definition: the number is cells that have intact membranes and took up the dye, which is much closer to what an experiment means by a cell. It stops at anything the dye cannot distinguish, notably a cell that is dying but not yet permeable, and at clumps, which a dye does not separate any better than an algorithm does. It also stops being comparable the moment the dye or the exposure changes, so record both in the method and treat a change of reagent lot the way you would a change of instrument.
A hemocytometer counting chamber is still the reference
The chamber is slow, operator dependent and the only method whose geometry is defined rather than calibrated. That makes it the right thing to check an instrument against when a number stops making sense, and the right thing for a small number of precious samples where the instrument's minimum volume is more than you have. Count enough squares that the statistics are real rather than hopeful, and have the same person do it when the comparison matters.
What a cell viability analyzer measures, and what it infers
Most instruments sold on viability measure membrane integrity with a dye and report the ratio, which is a proxy rather than a measurement of whether a cell will grow. A culture can read high on that ratio and fail to recover, because membrane integrity is the last thing to go. Where viability matters commercially or clinically, pair the dye with a metabolic or a plating readout at least once to see how the proxy behaves on your material, and then use the fast instrument knowing what it stands for.
Which flow cytometry viability dye to pair with it
On a cytometer the choice is between a membrane impermeant dye read in a free channel and a fixable dye that survives permeabilisation for an intracellular panel. The first is cheaper and adequate for a surface panel; the second is the only option when the cells will be fixed, and it has to be added before the fixative rather than after. Whichever it is, keep it in a channel that does not crowd a scarce marker, because dead cells bind antibodies non specifically and the viability gate is what removes them.
Holding one instrument and one protocol for a study
The largest source of disagreement in counting is not the instrument, it is the dilution, the mixing and the time between them. A suspension left for five minutes settles measurably, a pipette drawn too fast shears clumps into fragments the counter then counts, and a dilution made once rather than in series carries its own error into everything downstream. Write the protocol down to the mixing, and where a number will be compared across sites, exchange a reference suspension rather than a specification.
Questions people ask about cell counting instruments
Why do two instruments give different counts on one sample?
Because they are counting different things, and both are internally consistent. A brightfield counter includes debris the size of a cell and excludes a cell in a clump; a dye based counter includes only stained cells. Decide which definition your experiment needs, then keep that instrument for the study rather than trying to reconcile the two.
Is an automated counter accurate enough for a growth curve?
Yes, and precision rather than accuracy is what a growth curve needs: a consistent bias cancels out of a rate. What ruins a curve is changing the dilution protocol or the instrument halfway, which is why the protocol matters more than the specification sheet.
Do I still need a chamber if I have an instrument?
Keep one and use it twice a year and whenever a number surprises you. It is the only method whose volume is defined by geometry, so it is the arbiter when an instrument drifts or a new sample type reads implausibly.